Toxicology and industrial Health, Vol. 1 I , No. 2, 1995 119

METHYL-TERTIARY-BUTYL ETHER (MTBE) - A GASOLINE ADDITIVE - CAUSES TESTICULAR AND LYMPHO-

HAEMATOPOIETIC CANCERS IN RATS

FIORELLA BELPOGGI, MORANDO SOFFRITTI, AND CESARE MALTONI

Bentivoglio Castle Cancer Research Center Ramazzini Foundation of Oncology and Environmental Saences

and Bologna Institute of Oncology

Bologna, Italy

In the framework of a series of experiments conducted to evaluate the carcinogenic effects of oxygenated gasoline additives, MTBE was analyzed in an oral lifetime carcinogenicity study using &week-old male and female Sprague- Dawley rats. These experiments were part of a large research project on gasoline carcinogenicity performed at the Bentivoglio (BT) Castle Cancer Research Center of the Ramauini Foundation and of the Bologna Institute of Oncology. MTBE, dissolved in oil, was administered by stomach tube at the doses of 1000, 250, or 0 mgkg b.w., once daily, four days weekly, for 104 weeks. The animals were maintained until natural death. The last animal died 166 weeks after the stan of the experiment, i.e., at 174 weeks of age. Under the tested experimental conditions, MTBE was shown to cause an increase in Leydig interstitial cell tumors of the testes and a dose-related increase in lymphomas and leukemias in female rats

INTRODUCTION

Methyl-tertiary-bury1 ether (MTBE) is one of the most widely produced synthetic compounds in the world. Its major, if not unique use, is as an oxygenated additive to gasoline intended to improve the combustion process and more specifically, to significantly reduce motor vehicle CO emission, especially during low temperatures in winter months.

1. Address all correspondence to: Professor Cesare Maltoni, Institute of Oncology, Vide Ercolani, 412,40138 Bologna, ITALY. Tel.: 511390417.

2. Abbreviations: BT, Bentivoglio; DIPE, di-isopropyl ether; GLP, good laboratory practices; MON, motor octane number; ETBE, ethyl-teniaty-butyl ether; MTBE, methyl-tertiary-butyl ether; RON, research octane number; SOP, standard operating procedures; TAME, tert-amyl-methyl ether.

3. Key words: leukemia, Ley* cell testplar tumor, q E , rat. 4. This research was'partly supported by the Medical Research Program of the Emilia-Romagna

Region (Italy), Project No. 551.

Toxicology and Industrial Health, Vol. 11, No. 2, pp. 119-149 Copyright 1995 Princeton Sdentlfic Publishing Co., Inc.

ISSN: W40-2337

120 Belpoggi et al .

MTBE (C5HlzO), which is produced by reacting isohutylene with methanol, is a clear,

colorless liquid, with a characteristic terpenic odor. Its structural formula is:

H3C- C-0- CH3

I The properties of the compound are shown on Table 1.

TABLE 1. Properties of MTBE ' Property Values

Molecular weight 88

Specific gravity (15.6"C) 0.747

Net energy content. Id/g 34.9

Reid vapor pressure, kPa 53.6

Boiling point. "C 55 Heat of vaporization, J/g 320

Octane ratingZ RON 117

MON 102

Water solubility at 21°C MTBE in water, vol. % 6.9

Water in MTBE, vol. % 1.4 ~~~~

Stoichiometric A/F 11.7 '~rorn CHEMTECH, 1979. 2~~~ research octane number; MON: motor octane number.

Presently, MTBE is produced worldwide at a rate of more than 16 million tons per year. It is mainly consumed in the United States, particularly in the cooler states. In these cooler states, it is added to gasoline at concentrations of up to 15%. During the period 1984-1988, the percent of MTBE-containing gasoline increased from 8 to 22% in these states. In 1995, MTBE use in the United States is estimated to average 283,900 bld, equivalent to 12.3 million tonslyear (Williams, 1995). In Italy, where the industrial production of MTBE hegan in 1985 and its commercial expansion as a gasoline optimizer began in 1992, 300 thousand tons are produced annually. Since MTBE is added to gasoline at such high concentrations, it must not

Toxicology and Industrial Health, Vol. 11, No. 2. 1995 121

be considered as only an additive, but a true basic component of oxygenated gasolines. The use of MTBE as a component of gasoline has been recommended on the assumption that it would ameliorate the quality of the combustion exhausts, in terms of environmental and health safety (API, 1988; U.S. EPA, 1988).

Data on the long-term health effects of MTBE were not available at the time its use as an additive in premium gasoline was recommended and initiated in the 1970s. It is astonishing that such a technological process could have been started without sufficient toxicological information that would have enabled us to expose possible adverse health effects of the compound.

In 1988, the production and use of MTBE was already significant. Further, its use was predicted to expand, not only domestically but worldwide. We therefore began an early experimental research study (long-term bioassays) to assess if, and to what extent, MTBE could pose carcinogenic risks. This research was planned to provide qualitative, primary information. The study would be further investigated through larger, long-term carcinogenicity bioassays with the correlated mechanistic study.

This study was a part of a vast series of experimental research investigations performed at the Cancer Research Center of the Ramazzini Foundation and of the Bologna Institute of Oncology, at the Bentivoglio (BT) Castle, aimed at studying the carcinogenic risk of oxygenated gasoline additives presently used and proposed as gasoline optimizers. The plan and the state-of-the-art of this series of experiments are given in Table 2. All the experiments were performed on Sprague-Dawley rats, kept under observation for life span, following good laboratory practices (GLP) and applying the same standard operating procedures (SOP).

This report deals with the results of our experiments on the carcinogenicity of MTBE. Preliminary results were given at the Collegium Ramazzini Conference on "PRESENT KNOWLEDGE ON THE HAZARDS OF CONVENTIONAL AND NEW GASOLINES" held in Carpi, Italy, on October 29, 1993, in the framework of the "Annual Ramazzini Days, 1993."

MATERIALS AND METHODS

The purity of the MTBE, supplied by a domestic (United States) gasoline company, was higher than 99%. During the experiment, it was stored at a temperature of 4'C.

Male and female Sprague-Dawley rats of the colony at the BT Cancer Research Center were used. This colony of rats has been used for various experiments in the BT Laboratory for nearly 25 years. Historical data are available on more than 10,000 historical controls, kept under observation for lifespan and submitted to systematic necropsies and standardized histopathological examinations. Therefore, data on the expected incidence of the different types of tumors in control animals and its fluctuations are available.

+

TABLE 2. BT Plan of the Series of Long-Term (Life-Span) Carcinogenicity Bioassays of Oxygenated Gasoline g Additives, and of Oxygenated Additive-Containing Gasolines, Performed on Sprague-Dawley ~ a t s '

hr - Animals

(Part 5 State of the art ar

Compound No. (and code) Expen- Treatment 2. of the mental

experimnt group % Age at E

Sex No. S M

(weeks)

Methyl alcohol I (BT1001) I 15 mgll Ingestion In drinking water M 50 7 Histopathologically supplied ad libitum for F 50 evaluated; results in press 104 weeks

11 0 (control) M 100 F 100

2 (BT 960) 1 20,000 mgll Ingestion In drinbing water M 100 8 Biophase concluded su~died ad libitum for F 100 164 weeks

M 100 F 100

3 (BT961) I 24 mgkg b.w Ingestion By stomach tube, M 60 108 weeks of biophase in alive oil, 4 days F 60 weekly, for 104 weeks

II d control M 60 F 60

'Exprimcots 3. 5, 7, and 10 have the conml group in common: expenmcnu 8 and 9 have the cooi~ol gmup in mmmon. %live oil alone.

Toxicology and Industrial Health, Vol. I I , No. 2, 1995 123

TABLE 2. BT Plan of the Series of Long-Term (Life-Span) Carcinogenicity Bioassays of Oxygenated Gasoline $ Additives, and of Oxygenated Additive-Containing Gasolines, Performed on Spmgne-Dawley ~ a t s ' 00

??.

(Pnrtm) E

Compound No. (and code) Expen- of the mental Treatment Animals State of the art

experiment group A%e af

Type of sfan Dose Route admiidration Sex No. (weeks)

-

MIBE 6 (BT 958) I 1.W m& Ingestion By stomach Nbe, in olive M 60 8 Results verbally presenled af the b.w oil, 4 days weekly. for F 60 Ramazzini Days in Carpi, Italy,

104 weekx Oclober 1993 -Thisreport

11 250 mg/Lg b.w.

111 d (control)

ETBE 7 (BT 959) I 1,000 mglkg Ingestion By stomach tube, la olive M 60 8 108 weeks of biophase b.w. oil, 4 days weekly, for F 60

104 weeks 11 250 mg/Lg M 60

b.w. F 60

111 d M 60 (conuol) F 60

'Erommnts 3. 5. 7, and 10 have tbe contra1 goup in cammom; experimots 8 and 9 have me cooml g m p in commoo.

TABLE 2. BT Plan of the Series of Long-Term (Life-Span) Carcinogenicity Bioassays of Oxygenated Gasoline Additives, and of Oxygenated Additive-Containing Gasolines, Performed on Sprague-Dawley ~ a t s '

Pall Iv)

Compound No. (and code) Experi- ofthe mental Tnatmeat Aaimals Stlfe of the art

experiment group Age at

Type of, . start Dose Route iadmmmhon Sex No. (weeks)

Y $:

8 12 w& of biophase S' TAME 8 (BT963) 1 750mg/lrg b.w, Ingestion By in olive stomach oil, 4 tube, days M F 100 100 % weeuy, for 104 weeks

2 11 2M mg/Lg b.w. M 100

F 100 : Ill d M 100 5

(control) F 1M) 5 3. %

DlPE 9 (BT964) I l , W mg/lrg Ingestion By stomach tube, M 100 8 12 weeks of biophase in olive oil, 4 days F 100

5 b.w. ft

weekly, for 104 .S weeks

I1 250 mgkg b.w. M 100 F I00

3 - L. L.

(control) P IW

' E z ~ m o s 3. 5 , 7, and 10 have ihc cooml group io comrmo; experimou 8, and 9 have the coorml grmp in common.

F

TABLE 2. BT Plan of the Series of Long-Term (Life-Span) Carcinogenicity Bioassags of Oxygenated Gasoline N m Additives, and of Oxygenated Additive-Containing Gasolines, Performed on Sprague~Dawley ~ n t s '

mV, ? %

No. (and code) of Experi- ai Compound the experiment mental Treatment Animals State of the art 00.

W U P m, Age at start P

Type of Dose Route admimsuation Sex No. (weeks)

I RM) mgkg Ingestion By stomachtube, in M 60 8 108 weeks of biophase Gasoline containing 10 (BT 961) 3% methyl alcohol b.w. olive ail, 4 days F 60

weekly, for 104 weeks

Gasoline containing 5% ethyl alcohol

Gasoline

I I (BT 962) I 800 mgkg Ingestion By stomach tube, in M 60 8 75 weeks of biophase @soline containing b.w. olive oil. 4 days F 60 15% MTBE weekly, for 104

weeks

Gasoline containing 15% ETBE

'&eriments 3. 5 . 7. and 10 have the conaol gmup in common; erpcnmenls 8 and 9 have ihc cooaol group io common.

I Toxicology andindustrial Health, Vol. 11, No. 2, 1995 127

After weaning, at 4-5 weeks of age, the experimental animals were identified by ear punch, randomized in order to have no more than one male and one female of each litter in the same group, and housed five per cage. The animals were eight weeks old at the start of the experiment.

The plan of the experiment is shown in Table 3.

t

TABLE 3. Long-Term Carcinogenicity Bioassays on Methyl-Tertiary-Butyl Ether (MTBE), Administered by Stomach Tube to Sprague-Dawley Rats (Experiment BT 958)

Plan of the Experiment

Group Dose Animals (mgkg b.w. in olive oil) Sex No.

1 1000 M 60 F 60

M+F 120

(control) F 60

'Olive oil alone.

Every single dose of MTBE was administered by gavage in 1 ml of extra virgin olive oil, once daily, four days weekly (Monday and Tuesday; Thursday and Friday), for 104 weeks. A daily administration of the compound at the highest dose would not have been tolerated by the rats. The solutions were prepared weekly and maintained at 4'C.

The animals were kept under observation until natural death, under highly standardized housing and diet conditions identical to those used in the BT Laboratory over the last 20 years. In the experiments performed at the BT Cancer Research Center, the animals are usually allowed to live until natural death. By doing so, it is possible, to a feasible extent, to develop all the neoplastic potentialities. Mean daily feed and drinking water consumption were determined once weekly for the first 13 weeks from the start of the experiment, then every 2 weeks, until 112 weeks of age. Individual animal weight was measured once weekly from 8 weeks of age for the first 13 weeks, then every 2 weeks until 112 weeks of age, and

128 Belpoggi et al.

every 8 weeks until the end of the experiment. In order to detect and register all gross lesions, the animals were examined every week for the first 13 weeks, and then every 2 weeks until the end of the experiment.

The biophase of the experiment terminated after 166 weeks, with the death of the last animal at the age of 174 weeks.

Upon death, the animals were submitted to systematic necropsy. Histopathology was routinely performed on skin and subcutaneous tissue, the brain, pituitary gland,.Zymbal glands, salivary glands, Harderian glands, cranium (five levels) (with oral and nasal cavities and external and internal ear ducts), tongue, thyroid and parathyroid, pharynx, larynx, thymus and mediastinal lymph nodes, trachea, lung and mainstream bronchi, heart, diaphragm, liver, spleen, pancreas, kidneys and adrenal glands, oesophagus, stomach (fore and glandular), intestine (four levels), bladder, prostate, uterus, gonads, interscapular fat pad, subcutaneous and mesenteric lymph nodes, and any other organ or tissue with pathological lesions.

All organs and tissues were immediately preserved in 70% ethyl alcohol, except bones which were fixed in 10% formalin and then decalcified wlth a 10% formaldehyde and 20% formic acid in water solution. The normal specimens were trimmed following the standard procedure of the BT Laboratory: i.e., parenchymal organs were trimmed to allow for the largest surface area possible, and hollow organs were dissected through the hilus. The pathological tissue was trimmed to allow for the largest surface with normal adjacent tissues. The trimmed specimens were processed as paraffin blocks, and 3-5 micron sections of every specimen were obtained. Sections were routinely stained with haematoxylin-eosin. Specific stainings were performed when needed. All slides were examined microscopically by the same group of pathologists; a senior pathologist reviewed all the tumors and any other lesion of ontological interest. All pathologists followed the same criteria of histopathological evaluation and classification.

Two statistical methods were used for the analysis of tumor incidence. The first method is a prevalence analysis for nonlethal tumors and is described in Hoe1 and Walburg (1972). The Leydig cell testicular tumors given in Table 5 were considered nonlethal. The second method assumes lethality of the tumor and the statistical method is a log-ranked test and is described in Mantel (1966) and Cox (1972). This was used for the lymphomasileukemias on Table 6.

RESULTS

There were no differences in water and feed consumption among the animals of the treated and control groups of both sexes (Graphs 14).

. , , . . . There were no relevant differences in mean body weights of treated and control group male and female rats (Graphs 5 and 6). The survival was the same among male rats of the treated and control groups up to 88 weeks of age (80 weeks from the start of the treatment); then there was a higher survival of the males treated at the higher dose (Graph 7). A dose-response

Toxicology andindustrial Health, Vol. 11, No. 2, 1995 129

decrease in survival was observed among female rats from 24 weeks of age (16 weeks from the start of treatment) (Graph 8).

No evident behavioral changes were observed in MTBE-treated animals.

No relevant nononcological changes were detected by gross inspection and histological examination.

The benign and malignant tumors observed among male and female rats are listed in Table 4

The incidence of testicular Leydig cell tumors increased sharply in rats treated with the higher dose (Table 5). The increase was statistically significant at the p=0.05 level.

. , ,...,.

MTBE causes a dose-related increase in lymphomas and leukemias among female rats (Table 6). The fluctuations in the expected incidence of these neoplasias among female Sprague- Dawley rats in our historical controls is currently in a range below 10%. The incidence of lymphomas and leukemias in male Sprague-Dawley rats, which is slightly lower among animals exposed to the higher dose of MTBE, is within the range of the expected fluctuations in the historical controls. The increase in the females was highly significant (p < 0.01) at the highest dose level and marginally significant @ < 0 . b ) at the mid-dose level. K

. , .'.

An increase in dysplastic proliferation of lymphoreticular tissue was found in female rats treated with MTBE at both doses (Table 7). The term dysplastic proliferation of lymphoreticular tissue defines a hyperplastic lymphoid tissue, at various body sites, in which atypical lymphoid cells (usually lymphoimmunoblasts), isolated andlor aggregated in small clusters, are observed. The incidence of this type of change is higher among animals treated at the lower dose. This apparently paradoxical effect is probably due to the fact that more of these dysplastic proliferations might have developed into lymphomas and leukemias among female rats treated at the higher dose.

The incidence of all the other tumors at various sites was in the range of the expected fluctuations. A dose-related decrease in mammary fibromas and fibroadenomas, and in pituitary adenoma and tumors of adrenal glands was observed in treated female animals; this effect may be due to the increased dose-related early mortality caused by the treatment with . .- MTBE since the incidence of these tumors is age-dependent in our colony of Sprague-Dawley rats. An increased incidence of uterine sarcomas was observed in female rats exposed to the lower dose.

CONCLUSIONS AND DISCUSSION

.. MTBE is rapidly absorbed following oral exposure (Bio-Research Laboratories Limited, . . . 1990a). In the tested experimental conditions, MTBE, administered by stomach tube, caused an increase in Leydig interstitial cell tumors of testes in male rats; in female. rats, an increase in lymphomas and leukemias was noted. Moreover, an increase in uterine sarcomas was found in female rats treated with the lower dose. Regarding this last finding, it is interesting to note

130 Belpoggi et al.

Toxicology and Indusmrial Health, Vol. 1 1 , No. 2, 1995 131

132 Belpoggi et a1 . : .

. . ; . ,

Toxicology andindustrial Health, Vol. 11, No. 2, 1995 133

e - z n

MEAN BODY WEIGHT IN MALE RATS

c MlEE loo0 mg/lCgbw *. MlEE 250 mg/ kg bw a - MlEE 0 (control)

Start o l the experiment

0 I

0 8 24 40 56 72 88 104 120 136 152 168 184

Age (weeks)

GRAPE 5. Long-term carcinogenicity bioassays on METHYL-TERTIARY-BUTYL ETHER (MTBE). administered by stomach tube to Sprague-Dawley rats (Experiment BT 958).

Toxicology and Industrial Health, Vol. 11, No. 2, 1995 135

136 Belpoggi et al.

Toxicology andlndustrial Health, Vol. 11, No. 2, 1995 137

9

TABLE 4. Long-Term Carcinogenicity Bioassays on METHYL-TERTIARY-BUTYL ETHER (MTBE), Administered by Stomach Tube to Sprague-Dawley Rats (Experiment BT 958) %' m

Incidence of Various Types of Benign and Malignant Tumors among Male Spragne-Dawley Rats ?i

(Part I) P

Site Histotype Groups

I: 1000 mglkg b.w. 11: 250 mgkg b.w. I l l 0' (control)

No. % No. 96 No. %

Subcutaneous tissue Fibroma 1 1.7 0 1 1.7 - Lipoma 1 1.7 0 - 0 - Liposarcoma 1 1.7 0 - 0 -

Mammary gland Fibroma 3 5.0 2 3.3 1 1.7

Fibrolipoma 0 - 1 1.7 0 -

Lipoma 0 - 0 - 1 1.7

Adenocarcinoma 1 1.7 1 1.7 1 1.7

Liposarcoma 0 - 0 - 1 1.7

Zymbal gland Carcinoma 2 3 .3 1 1.7 3 5.0

Ear dud Carcinoma 3 5.0 3 5.0 2 3.3

Nasal cavities Olfactory neuroblastoma 0 - 0 - 1 1.7 -- Oral cavity, lips, and tongue Carcinoma 2 3.3 1 1.7 0 -

'Olive oil alone.

TABLE 4. Long-Term Carcinogenicity Bioassays on METHYL-TERTIARY-BUTYL ETHER (MTBE), Administered by Stomach Tube to Sprague-Dawley Rats (Experiment BT 958)

Incidence of Various Types of Benign and Malignant Tumors among Male Sprague-Dawley Rats

Site H~stotype Groups

I: 1000 mf lg b.w. 11: 250 mglka b.w. 111: 0' (control)

No. % No. % No. %

Squamous cell carcinoma 0 - 1 1.7 0 - Y

Larynx a 5 . ?.

Lung Fibroangioma Adenocarcinoma

Liver Fibroangioma

Hepatocarcinoma

Pancreas Islet cell adenoma 5 8.3 2 3.3 4 6.7 3 Exocrine adenocarcinoma 1 1.7 0 - 0 - % .s

Kidneys Adenoma 2 3.3 1 1.7 1 1.7 $ -- L.

Prostate Adenocarcinoma 0 - 1 1.7 0 - Z P N

TestesZ Leydig cell tumor 11 18.3 2 3.3 2 3.3 G 'olive oil alone. z Z ~ e e Table 5.

+ e 0

TABLE 4. Long-Term Carcinogenicity Bioassays on METHYL-TERTIARY-BUTYL ETHER (MTBE), Administered by Stomach Tube to Sprague-Dawley Rats (Experiment BT 958) g

'P 2

Incidence of Various Types of Benign and Malignant Tumors Among Male Sprague-Dawley Rats 3. m

(Part m) F- Site Histotype G r o u p s

I: IOOO m a g b.w. 11: 250 m$g b.w. 111: 0' (control)

NO. % NO. % NO. %

Pituitary gland Adenoma 10 16.7 9 15.0 11 18.3

Thyroid gland C-cell adenoma 1 1.7 2 3.3 4 6.7

C-cell carcinoma 1 1.7 0 - 0 -

Adrenal glands Cortical adenoma 0 1 1.7 1 1.7 -

Pheocromocytoma 20 33.3 13 21.7 13 21.7

Pheocromoblastoma 1 1.7 1 1.7 0 -

Central nervous system - Brain Oligodendroglioma 3 5.0 2 3.3 3 5.0

Neuroblastoma 0 - 1 1.7 0 -

Meningioma 0 - 1 1.7 0 - - Meninges

Bones -Cranium Osteosarcoma 1 1.7 2 3 , 3 3 5.0 'Olive oil alone.

TABLE 4. Long-Term Carcinogenicity Bioassays on METHYL-TERTIARY-BUTYL ETHER (MTBE), Administered by Stomach Tube to Sprague-Dawley Rats (Experiment BT 958)

Incidence of Various Types of Benign and Malignant Tumors among Male Sprague-Dawley Rats

(Part IV) Site Hlstotype Groups 3 2.

I: 1000 mgkp b.w. 11: 250 m a g b.w. m: 0' (control) 2 No. % No. % No. %

Soft tissues Liposarcoma 0 - 1 1.7 0 - f: a

Heart Malignant Schwamoma 0 - 0 - 1 1.7 s F 3 a.

Subcutaneous lymph nodes Fibroangioma 0 - 1 1.7 0 - FL 3 a_

Haemolymphoreticular Lymphomas and leukemias 7 11.7 9 15.0 10 16.7 -% ~ - ~.

tissue^^.^ s 'Olive oil alone. ?-

31ncluding subcutaneous lymph nodes. 4See Table 6.

- TABLE 4. Long-Term Carcinogenicity Bioassays on METHYL-TERTIARY-BUTYL ETHER (MTBE), p,

Administered by Stomach Tube to Spmgue-Dawley Rats (Experiment BT 958) F

Incidence of Various Types of Benign and Malignant Tumors among Female Sprague-Dawley Rats 011

00. (Pan V)

Site Histotype Groups f I: 1000 mg/kgb.w. 11: 250 me/ke b.w. 111: 0' (control)

No. % No. % No. %

Skin Dermatofibrollla 0 - 1 1.7 1 1.7

Squamous cell carcinoma 0 - 0 - 1 1.7

Lipoma 1 1.7 0 - 0 - Subcutaneous tissue

Mammary gland Fibroma and fibroadenoma 16 26.7 27 45.0 40 66.7

Fibrolipoma 1 1.7 0 - 0 -

Adenocarcinoma 2 3 . 3 6 10.0 4 6.7

Carcinosarcoma 0 - 1 1.7 0 -

Zymbal gland Carcinoma 2 3 . 3 0 2 3.3 -

Ear duct Acanthoma Carcinoma

Oral cavity, lips, and tongue Acanthoma 1 1.7 0 - 2 3 . 3 Odontoma 0 - 0 I 1.7 - Carcinoma 0 - 2 3 . 3 1 1.7

'olive oil alone.

TABLE 4. Long-Term Carcinogenicity Bioassays on METHYL-TERTIARY-BUTYL ETHER (MTBE), Administered by Stomach Tube to Sprague-Dawley Rats (Experiment BT 958)

Incidence of Various Types of Benign and Malignant Tumors among Female Sprague-Dawley Rats

(Part VO Site Histotype Groups

1: 10M) mgkg b.w. 11: 250 mg/kg b.w. 111: 0' (control)

No. % No. % No. %

Larynx Squamous cell carcinoma 0 - 1 1.7 0 - Y 8. 5

Pharynx Squamous cell carcinoma 1 1.7 0 - 0 - F 3

Lung Fibroangioma

Stomach

- Forestomach Acanthoma

- - g Liver Cholangiofibroma 0 I 1.7 0 .>

Pancreas Exocrine adenoma 1 1 .7 2 3.3 1 1.7 $ 4

Islet cell adenoma 2 3.3 I 1.7 I 1.7 L

2 9

Kidneys Lipoma 0 - 0 - 1 1.7 .u

' Olive oil alone. 4 '0

2

TABLE 4. Long-Term Carcinogenicity Bioassays on METHYL-TERTIARY-BUTYL ETHER (MTBE), - a

Administered by Stomach Tube to Sprague-Dawley Rats (Experiment BT 958) a

Incidence of Various Types of Benign and Malignant Tumors among Female Sprague-Dawley Rats F %- m

(Part- 00. n

Site Histotype Groups % I: 1000 m a g b.w. 11: 250 mgka b.w. Ill: 0' (control)

No. % No. % No. %

OVW Granulosa cell hunor 1 1.7 3 5.0 0 - Granulosa and theca cell tumor 0 - 0 - 1 1.7

Fibroma I 1.7 0 - 0 -

Uterus polyp 12 20.0 12 20.0 13 21.7

Fibroma 0 - 2 3.3 1 1.7 Fibroangioma 1 1.7 0 - 0 - Granular cell tumor 0 - 2 3.3 0 - (Abrikossoff s tumor)

Sarcoma 0 - 5 8.3 I 1.7

Uterus and vagina Sarcoma 0 - 0 1 1.72 -

Peritoneum Lipoma 0 - 0 - 3 5.0

Pituitary gland Adenoma 13 21.6 16 26.7 22 36.7

Thvmid eland Follicular cell adenoma 2 3.3 0 - 0 - C-cell adenoma 3 5.0 10 16.7 5 8.3

'Olive oil alone. This sarcoma was involving both uterus and vagina.

TABLE 4. Long-Term Carcinogenicity Bioassays on METHYL-TERTIARY-BUTYL ETHER (MTBE), Administered by Stomacb Tnbe to Sprague-Dawley Rats (Experiment BT 958)

Incidence of Various Types of Benign and Malignant Tumors among Female Sprague-Dawley Rats

Site Histotype Groups

I: 1000 mflg b.w. 11: 250 mp/kg b.w. m: 0' (control)

Adrenal glands Cortical adenoma 3 5.0 6 10.0 5 8 .3 Pheocromocytoma 10 16.7 11 18.3 18 30.0 Cortical adenocarcinoma 0 - 4 6.7 3 5.0 Pheocrornoblastoma 0 - 0 - 2 3 . 3

Central nervous system - Brain Oligodendroglioma 0 - 0 - 1 1.7 - Meninges Meningioma 0 - 1 1.7 0 -

Bones -Cranium Osteosarcoma

Mesenteric lymph nodes Fibroangioma 0 - 1 1.7 0 -

Haemolymphoreticular t i s s u e ~ ~ . ~ Lymphoma and leukemia 12 20.0 6 10.0 2 3 . 3

'olive oil alone. 31ncluding mesenteric lymph nodes. 4See Tables 6 and 7.

146 Belpoggi et al.

that an increase in these tumors was found in the same colony of Sprague-Dawley rats following treatment with gasoline (with high content of aromatics) (Makoni et al., 1995). On the basis of our results, MTBE must be considered an animal carcinogen, and therefore represents a potential carcinogenic risk for humans.

TABLE 5. Long-Term Carcinogenicity Bioassays on Methyl-Tertiary-Butyl Ether (MTBE), Administered by Stomach Tnbe to Sprngue-Dawley Rats (Experiment BT 958)

Results: Leydig Cell Testicular Tumors Dose

G~~~~~ (mglkg b.w. No. at Corrected Animals bearing NO. in olive oil) start number' No. Leydip, cell %2 tumors %3

32 11 34.44 I 1M)o 60

25 2 2.3 8.0 II 250 60

26 2 2.3 7.7 ID o5 60 (control)

male rats at 96 weeks of age, when the first Leydig cell tumor was observed. ?-Percentages are referred to the number at start. 3Percentages are referred to the corrected number. 4p c 0.05. solive oil al0ne.p

TABLE 6. Long-Term Carcinogenicity Bioassays on Methyl-Tertiary-Bntyl Ether (MTBE), Administered by Stomach Tube to Sprague-Dawley Rats (Experiment

Results: Lym homas and Leukemias Dose

Groups (mg/kg b.lu. Male Female NO. in ohve 011)

iyo, corrected .~.~imals bearing No. Corrected Animals bearing at number' Lymphomas and stui at numbe? Lymphomas

and S t a n Lediemias Lcukemiss

No. %3 NO. %3 4b4 1 loo0 60 58 7 11.7 12.0 60 47 12 25.5'

I1 250 60 59 9 15.0 15.3 60 51 6 IlLQ 11.8~

59 10 16.7 16.9 60 58 2 22 3.4 m 0' 60 (conuol)

1,b.live male rats at 32 weeks of age, when the fust leukemia was observed. ZAlivc female rats at 56 weeks of age, when the fust leukemia was obsetved. 'percentages are referred to the number at stui. "percentages are r e f e d to the conected number. 'p <0.01. 6p < 0:01. 'Olive oil alone.

Toxicology and Industrial Health, Vol. 11. No. 2, 1995 147

. . . . ._>

TABLE 7. Long-Term Carcinogenicity Bioassays on Methyl-Tertiary-Butyl Ether (MTBE), Administered by Stomach Tube to Sprague-Dawley Rats

, . , . , . '..

(Experiment BT 958) ., , .,

Results: Dysplastic Proliferation of Lymphoreticular Tissues (DPLT) at Various Sites

Dose Groups (mglkg b.w. Animals Animals bearing DPLT No. in olive oil)

Sex No. at Corrected No. %2 ?b3 start number1

I 1000 F 60 59 9 liS 15.3

111 o4 F 60 60 1 El 1.7

(control) 'Alive female rats at 26 weeks of age, when the first DPLT was observed.

Percentages are referred to the number at start. ' Percentages are referred to the corrected number. 'Olive oil alone.

In another long-term carcinogenicity study (known as the ARCO Study), supported in the United States by a group of producers and users of MTBE, MTBE was tested on both male and female Fischer 344 rats and CD-1 mice. In this study, the compound was tested by inhalation at concentrations of 8000, 3000, 400, and 0 ppm. The results of this unpublished study showed MTBE to be carcinogenic (Burleigh-Flayer et al., 1992; Chun et a]., 1992); the compound increased the incidence of a rare type of kidney tumor (renal tubular adenomas and carcinomas) in male rats exposed to high-dose and particularly to mid-dose. This was statistically significant when compared to concurrent controls. Further, a mid-dose-related increase (still statistically significant) in "interstitial cell adenoma of testes" was noted in male rats. This increase was difficult to evaluate because of the high background of the spontaneous incidence of those tumors in their historical controls (Haseman et a]., 1985). The results of the ARCO study also showed that MTBE causes an increase in the incidence of hepatocellular adenomas in female mice. In the ARCO rat study there was a high and apparently dose-related incidence of noncancer renal toxicity in both sexes, an effect which was not observed in our experiment.

It is interesting to note that in the studies conducted by ARCO and at the BT Laboratory, MTBE was found to increase the incidence of Leydig cell testicular tumors. Our results and the results of the ARCO study, although different, converge in exposing the carcinogenic potential of MTBE. Moreover, the combined results of the two studies appear to indicate that MTBE is a trans-species, multistrain, multisite carcinogen.

148 Belpoggi et a1

The most striking indication of MTBE carcinogenicity is the increase in lymphomas and leukemias among female rats observed in the BT bioassays. There are not yet sufficient data to define the mechanisms by which MTBE acts as a leukemogen, although some available information may prove helpful. MTBE was reported to metabolize to tertiary butyl alcohol and formaldehyde by microsomal demethylation (Savolainen et al., 1985; Brady et al., 1990). On the other hand, preliminary results of long-term carcinogenicity bioassays of formaldehyde administered by drinking water showed that this compound increases the incidence of lymphomas and leukemias in both males and females (Soffritti et al., 1989). These early results were more consistently confirmed by our complete and conclusive data (Soffritti et al., 1995). Pharmacokinetic studies have shown sex differences which deserve further investigation; after exposure, a decreased plasma concentration of MTBE was observed in female rats (Bio-Research Laboratories Limited, 1990b). This result suggests that in females, MTBE undergoes a more rapidlintense metabolization. All these data seem to point to the formaldehyde transformation of MTBE as a probable crucial step in its leukemogenic effect.

On the basis of these results of the BT experiments, and of the other available information, MTBE must be considered as a potential human carcinogen. Therefore, its use constitutes a worldwide public health problem. This problem is aggravated by the fact that the addition of MTBE to gasoline produces increased levels of formaldehyde in the exhaust (Stump et al., 1990).

Further investigation is needed to better define and quantitatively assess the carcinogenicity of MTBE. Even today, however, the use of MTBE in gasoline must be reconsidered in view of the available data. Measures of prevention must be undertaken in the scenarios of heavier exposure

Additionally, in view of the fact that MTBE is found in the drinking water of some localities, this study provides a better assessment of risk by exposure in drinking water than the unpublished inhalation studies. Thus, the risk to public health from water contaminated by MTBE should be alleviated at once.

We wish to thank Dr. Myron A. Mehlman for having made available to us the tested MTBE. We are indebted to Dr. David G . Hoe1 for having performed the statistical analysis of the results dealing with the incidence of Leydig cell testicular tumors and lymphomasfleukemias.

REFERENCES

AMERICAN PETROLEUM INSTITUTE (API) (1988). "Alcohols and ethers: a technical assessment of their application as fuels and fuel components." 2nd ed. API publication 4261. American Petroleum lnstitute. Washington, DC

Toxicology and Industrial Health, Vol. I I , No. 2, 1995 149

BID-RESEARCH LABORATORIES LIMITED (1990a). "Mass balance of radioactivity and metabolism of methyl -tertiary-butyl ether (MTBE) in male and female Fischer-344 rats after intravenous, oral and dermal administTation of 14C-MTBE." Bio-Research, Senneville, Qukbec. Canada.

BIO-RESEARCH LABORATORIES LIMITED (1990b). Pharmacokinetics of Methyl-Tertiary-ButyI Ether (MTBE) and tertiary-butyl alcohol (TBA) in male and female Fischer-344 rats after single and reveat inhalation nose-only exoosures to MTBE. Reoon No. 38844. Bio-Research Laboratories id., Senneville, Quebec, danaia.

BRADY, J.F., XIAO, F., NING. S.M., and CHUNG, S.Y. (1990). "Metabolism of methyl-tertiary-bum1 . . ether by rat hepatic microsomes." Arch. Toxicol. 64:157-160.

BURLEIGH-FLAYER, H.D., CHUN, J.S., and KINTISH. W.J. (1992). Methyl-Tertiary-Butyl Ether: vapor inhalation oncogenicity study in CD-1 mice. BRRC repon 91NW13A. Union Carbide, Bushy Run Research Center, Expon, PA.

C m h m c ~ (1979). August:502-510. CHUN, J.S.. BURLEIGH-FLAYER, H.D.. and KINTISH, W.J. (1992). Methyl-Tertiary-Butyl Ether:

vaoor inhalation oncoeenicitv studv in Fischer-344 rats. BRRC reoort 91N0013B. Union Carbide. ~ ; s h ~ Run Research 6 e n t e r , ~ x p o ~ , PA.

COX. D.R. (1972). "Remession models and lifetables (with discussion)." J.R. Stat. Soc. B 24:187-220. HASEMAN', J.K:, H*, J.E., RAO, G.N., ARNOLD, J.E., BOORMAN, G.A., and McCONNELL,

E.E. (1985). "Nwplasms observed in untreated and corn oil gavage control groups of F344N rats and (C57BL16N x C3WHeN) F1 (B6C3F1) mice." J. Natl. Cancer Inst. 75:975-984.

HOEL, D.G. and WALBURG, H.E. (1972). "Statistical analysis of survival experiments." 1. Natl. Cancer Inst. 49:361-372.

MANTEL, N. (1966). "Evaluation of survival data on two new rank order statistics arising in its consideration." Cancer Chemother. Rep. 50:163-170.

MALTONI, C., PINTO, C., CILBERTI, A.L., BELPOGGI, F., and SOFPRIlTI, M. (1995). "Long-term carcinogenicity bioassays on gasoline with high levels of aromatics, performed on Sprague-Dawley rats." In press.

SAVOLAINEN, H., PFAFFLI, P., and ELOVAARA, E. (1985). "Biochemical effects of methyl tertiary-butyl ether in extended vapour exposure of rats." Arch. Toxicol. 57:285-288.

SOFFRI?TI. M.. MALTONI. C.. MAFFEI. F.. and BIAGI. R. (1989). "Formaldehvde: an ex~erimental . . multipotential carcinogei."~ox.icol lid. Health 5:69&730.

SOFFRITTI. M.. BELPOGGI. F.. CILIBERTI. A.. BORTOLUZZI. L., SANTI. A., KALOGEROPOULOS, T.. and M ~ T O N I , C. (1995). ' ' ~on~- t e rm carcinogenicity bioassays on formaldehyde and acetaldehyde administered with drinking water: results and implications." Communication at the International Conference on "Drinking waters, other beverages, and their containers; a present problem on public health," held in Bologna on October 28 1994. In press.

STUMP, F.D., KNAPP, K.T., and RAY, W.D. (1990). "Seasonal impact of blending oxygenated organics with gasoline on motor vehicle tailpipe and evaporative emissions." I . Air Waste Manage. Assoc. 40.872-880.

UNlTED STATES ENVIRONMENTAL PRoTECnON AGENCY (U.S. EPA) (1988). Guidance on Estimating Motor Vehicle Emission Reductions from the Uuse of Alternative Fuels Blends. EPA technical report no. EPA-AA-TSS-PA-87-4. Office of Mobile Sources, Ann Arbor. MI.

WILLIAMS, B. (1995). "MTBE, ethanol advocates' squabble may complicate RFG implementation." OilGas I., Feb. 13:17-22. j . .